Improved molecular recognition of Carbonic Anhydrase IX by polypeptide conjugation to acetazolamide

Article abstract of DOI:10.1016/j.bmc.2017.09.017

The small molecule inhibitor acetazolamide (AZM) was conjugated to a set of designed polypeptides and the resulting conjugates were evaluated for their affinity to Human Carbonic Anhydrase II (HCA II) using surface plasmon resonance. The dissociation constant of the AZM-HCA II complex was 38 nM and that of the AZM conjugated polypeptide (4-C10L17-AZM) to HCA II was found to be 4 nM, an affinity enhancement of a factor of 10 due to polypeptide conjugation. For Human Carbonic Anhydrase IX (HCA IX) the dissociation constant of AZM was 3 nM, whereas that of the 4-C10L17-AZM conjugate was 90 pM, a 33-fold affinity enhancement. This dramatic affinity increase due to polypeptide conjugation was achieved for a small molecule ligand with an already high affinity to the target protein. This supports the concept that enhancements due to polypeptide conjugation are not limited to small molecule ligands that bind proteins in the mM to μM range but may be used also for nM ligands to provide recognition elements with dissociation constants in the pM range. Evaluations of two HCA IX constructs that do not carry the proteoglycan (PG) domain did not show significant affinity differences between AZM and the polypeptide conjugate, providing evidence that the improved binding of 4-C10L17-AZM to HCA IX emanated from interactions between the polypeptide segment and the PG domain found only in one carbonic anhydrase, HCA IX.

Full text of DOI:10.1016/j.bmc.2017.09.017

Accepted Manuscript
Improved molecular recognition of Carbonic Anhydrase IX by polypeptide
conjugation to acetazolamide
Jie Yang, Katarina Koruza, Zoë Fisher, Wolfgang Knecht, Lars Baltzer
PII:
S0968-0896(17)31344-5
DOI:
http://dx.doi.org/10.1016/j.bmc.2017.09.017
BMC 13977
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
30 June 2017
10 September 2017
13 September 2017
Please cite this article as: Yang, J., Koruza, K., Fisher, Z., Knecht, W., Baltzer, L., Improved molecular recognition
of Carbonic Anhydrase IX by polypeptide conjugation to acetazolamide, Bioorganic & Medicinal Chemistry (2017),
doi: http://dx.doi.org/10.1016/j.bmc.2017.09.017
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Improved molecular recognition of Carbonic
Anhydrase IX by polypeptide conjugation to
acetazolamide
The conjugation of acetazolamide to
a designed polypeptide increases its
a
b
b,c
b*
a*
Jie Yang , Katarina Koruza , Zoë Fisher , Wolfgang Knecht , Lars Baltzer
Department of Chemistry-BMC, Uppsala University, Husargatan 3, 752 37 Uppsala, Sweden; Department of Biology & Lund Protein Production
a
b
c
Platform (LP3), Lund University, Sölvegatan 35, 22362 Lund, Sweden; Scientific Activities Division, European Spallation Source ERIC (ESS),
Tunavägen 24, 22100 Lund, Sweden.

Bioorganic & Medicinal Chemistry
journal homepage: www.els evier.com
Improved molecular recognition of Carbonic Anhydrase IX by polypeptide
conjugation to acetazolamide
a
b
b,c
b,
a,
Jie Yang , Katarina Koruza , Zoë Fisher , Wolfgang Knecht ∗ and Lars Baltzer *
a
Department of Chemistry-BMC, Uppsala University, Husargatan 3, 752 37 Uppsala, Sweden
b
Department of Biology & Lund Protein Production Platform (LP3), Lund University, Sölvegatan 35, 22362 Lund, Sweden
c
Scientific Activities Division, European Spallation Source ERIC (ESS), Tunavägen 24, 22100 Lund, Sweden.
ARTICLE INFO
ABSTRACT
Article history:
Received
Received in revised form
Accepted
The small molecule inhibitor acetazolamide (AZM) was conjugated to a set of designed
polypeptides and the resulting conjugates were evaluated for their affinity to human Carbonic
Anhydrase II (HCA II) using surface plasmon resonance. The dissociation constant of the AZM-
HCA II complex was 38 nM and that of the AZM conjugated polypeptide (4-C10L17-AZM) to
HCA II was found to be 4 nM, an affinity enhancement of a factor of 10 due to polypeptide
conjugation. For Human Carbonic Anhydrase IX (HCA IX) the dissociation constant of AZM
was 3 nM, whereas that of the 4-C10L17-AZM conjugate was 90 pM, a 33-fold affinity
enhancement. This dramatic affinity increase due to polypeptide conjugation was achieved for a
small molecule ligand with an already high affinity to the target protein. This supports the
concept that enhancements due to polypeptide conjugation are not limited to small molecule
ligands that bind proteins in the mM to µM range but may be used also for nM ligands to
provide recognition elements with dissociation constants in the pM range. Evaluations of two
HCA IX constructs that do not carry the proteoglycan (PG) domain did not show significant
affinity differences between AZM and the polypeptide conjugate, providing evidence that the
improved binding of 4-C10L17-AZM to HCA IX emanated from interactions between the
polypeptide segment and the PG domain found only in one carbonic anhydrase, HCA IX.
Available online
Keywords:
Human carbonic anhydrase
Acetazolamide
Recognition
Surface plasmon resonance
Tumor
2
009 Elsevier Ltd. All rights reserved.
1
. Introduction
The specific high-affinity molecular recognition of proteins is
of association of the polypeptide segment. Since the small
molecule selected for conjugation binds to the protein, and as the
polypeptide is covalently linked to the small molecule, the loss of
translational degrees of freedom upon binding is minimized.
Interactions between polypeptide amino acid residues and those
of the target proteins can provide affinity enhancements of several
orders of magnitude in spite of the fact that independently they do
not bind to the protein in a measurable way. The role of
the key to a large number of applications in biomedicine and
biotechnology. The need for small organic molecules, peptides,
recombinant proteins and antibodies in diagnostic and therapeutic
applications continue to drive development of synthesis methods
and recombinant technologies. Such endeavours are labour
intensive and costly and concepts designed to increase affinities
and selectivities with less effort are of considerable interest. We
reported previously on a strategy based on the conjugation of
small organic molecules and peptides to a designed set of
polypeptides that for a variety of proteins gave rise to conjugates
hydrophobic contacts in generating affinity between the
polypeptide and the protein has been demonstrated in the case of
5
glycogen phosphorylase A. The charged residues next to the
hydrophobic ones in the sequence can form hydrogen bonds and
electrostatic interactions to provide further affinity enhancement
(
Figure 1) with affinities that in comparison to the small
3
but mainly increase selectivity between proteins. The concept of
molecules and peptides alone were enhanced by 3-4 orders of
tailing has been introduced, previously, to describe the extension
of small organic inhibitors of carbonic anhydrases by organic
groups, accessing amino acid residues further out in the binding
pocket, as compared to interactions in close proximity to the
1
-3
magnitude. The ability to discriminate between the highly
homologous Human Carbonic Anhydrases I and II was increased
by a factor of 30, measured by ratios of dissociation constants.4
The reason that polypeptides enhance affinities and selectivities in
6
catalytically active metal ion. Polypeptide conjugation extends
—
spit—e o—f weak interactions is due to the more favourable entropy
∗
Corresponding author. E-mail addresses: Wolfgang.Knecht@biol.lu.se (Wolfgang Knecht); lars.baltzer@kemi.uu.se (Lars Baltzer).

the area of interaction to residues outside of the binding pocket,
sampling non-obvious binding sites. The peptidic character also
affects biodistribution.
Human Carbonic Anhydrases (HCAs) catalyze the reversible
-
2 3
interconversion of CO to HCO , with a subsequent proton
transfer step that completes the catalytic cycle. There are 15
expressed HCA enzymes identified to date and they share 30-80%
sequence identity. They can be cytosolic, membrane-bound,
mitochondrial, extracellular or secreted and are involved with
numerous physiological functions, from cerebral spinal fluid
O
N
N
O
2
H N
S
S
N
H
O
AZM
O
S
N
N
O
13
O
S
N
N
O
production to bone remodeling. HCA I and HCA II are present
H
2
N
H
N
S
N
H
H2N
O
7
O
S
N
H
at high concentrations in the red blood cells and are well-
understood enzymes that are commercially available at modest
cost. Several small molecule inhibitors of HCAI and HCA II are
7
O
1
O
O
F
F
F
F
1
4,15
known.
F
Polypeptide conjugates are not expected to pass through cell
membranes which would exclude access in blood to HCA I and
HCA II that reside in the red blood cells at very high
Figure 1. Illustration of the polypeptide conjugate concept. A small
molecule, such as acetazolamide (AZM), a known inhibitor of carbonic
anhydrases, is modified by organic synthesis to incorporate a linker and a
functional group for conjugation to amino acid side chains to form active ester
concentrations. It is therefore of interest to investigate also the
affinity for extracellularly accessible HCA isoforms. HCA IX is a
multi-domain, glycosylated, dimeric membrane protein with the
catalytic CA domain presented to the extracellular side (see SI,
1
. The active ester is reacted with the amino group of a Lys side chain to form
the polypeptide conjugate, which binds to the target carbonic anhydrase
through the combined interactions of AZM and the polypeptide with an
affinity that is higher than that of AZM. The peptide part of the polypeptide
conjugate is here depicted as two ordered alpha helices for simplicity.
However the flexibility of the polypeptide structure allows it to adapt to a
variety of protein surfaces and thus to enhance affinities for a large number of
1
6
Figures S1; S2). Its presence is indicative of metastasis and acts
as a tumor independent marker of poor prognosis as recently
1
7
reported by a meta-analysis combining 147 independent studies.
Additionally, knockdown of CA IX and pharmacological
intervention with CA IX in breast cancer mouse models led to
smaller tumors, slower tumor growth, and resulted in reduced
2
-5, 30-32
proteins but alone it may show a high variety of confirmation states.
1
8,19
metastasis.
Similar effects of CA IX knockdown were found
20
The availability of small organic molecules that bind proteins
with dissociation constants in the nM range suggests that, in
principle, the development of molecules with pM affinities should
be feasible. Such molecules would compare well with the
properties of recombinant proteins in terms of affinity, selectivity
and robustness, while having molecular weights of 5 kD or less.
They would be expected to have diverse applications including
medical imaging, diagnostics of low abundance biomarkers in the
clinic as well as in point of care applications carried out by non-
specialists. There is furthermore an abundance of small organic
molecules and peptides that do not meet performance criteria with
regards to affinities and selectivities to target proteins and that
could be "rescued" and used for their original purposes or for
other purposes.
in a colon cancer model. There is therefore strong evidence that
identifying patients with high HCA IX expression and
subsequently treating them with HCA IX inhibitors could be
beneficial. However, there is an unmet medical need to develop
HCA IX specific probes or inhibitors for use in tumor imaging,
2
1,22,23
cancer detection and staging, and as therapeutic drugs.
The
problem with clinically-used sulfonamide-based drugs against
HCAs is that they do not show isozyme discrimination and bind
to off-target HCAs, causing harmful side effects. To specifically
target cancers that are HCA IX-positive there is a need for HCA
2
1,24
IX isozyme-specific drugs.
A
It is not self-evident that it is possible to enhance affinities also
of small molecules with already very high affinities. Even though
care would be exercised to incorporate the linker in a position not
directly involved in the interactions with the protein target it is in
fact a substantial structural modification of the small molecule.
High affinities of small molecules for proteins are associated with
multiple precisely organized and synergistic interactions between
the small molecule functional groups and those of the amino acid
side chains, backbone amide groups and also solvent molecules
and the perturbation of this finely tuned complexation might
B
7
,8
reduce the overall affinity.
In order to provide further understanding of the limitations and
boundaries of polypeptide conjugation the current study was
performed to determine if a small organic molecule with affinity
for a protein in the low nM range could be used to form
polypeptide conjugates with improved affinity. As target system
the carbonic anhydrases were selected as these model proteins are
well-known in their structure-function relationships, and several
small molecule inhibitors are already available over a range of
Figure 2. A, Left: Crystal structure of HCA II in complex with
acetazolamide (AZM) (PDB ID 4G0C). Right: crystal structure of native HCA
2
5,16
IX in complex with AZM (PBD ID 3IAI).
B: Surface representation of
9
affinities. This class of enzymes is also of considerable interest
native HCA IX, hydrophilic area in red, hydrophobic in green, residues in
parenthesis are found in HCA II.
to other groups, as shown by recent examples were carbonic
anhydrase inhibitors were conjugated to appropriate scaffolds to
increase affinity or carry a therapeutically relevant load to the
1
0-12
target enzyme.
Acetazolamide (AZM) is a known small molecule inhibitor of
the HCAs with dissociation constants in the low nM range.14

AZM does not discriminate strongly between carbonic anhydrases reaction, crude 2 was precipitated out from the aqueous solution
since interactions between the sulphonamide amino group and the
active site Zn ion dominate the binding (Figure 2).
upon neutralization with aqueous NaOH. After recrystallization in
ethanol, pure 2 was obtained in 51% yield. In the amidation step
b, methyl 10-chloro-10-oxadecanoate was reacted with the amino
group of 2 in the presence of pyridine. The reaction mixture was
acidified with 2 M HCl to precipitate the crude methyl ester 3,
and pure 3 was obtained in 63% yield after recrystallization. In
step c, the methyl ester 3 was hydrolyzed in 5% aqueous NaOH.
Upon carefully adjusting the pH to 7, the acid 4 was precipitated
and pure 4 was obtained in 51% yield after recrystallization. In
step d, an excess amount of pentafluorophenol compared to acid 4
was used to afford active ester 1, the reaction was promoted by
EDCI as a dehydration reagent. There was no purification step
since EDCI and its reacted form will go to the aqueous layer
during partitioning, and the excess amount of pentafluorophenol
was easily removed under vacuum overnight. Characterization of
The high affinity of AZM and the further enhancement of
affinity and selectivity expected from polypeptide conjugation,
suggested that inhibitors with considerably improved molecular
properties over those of AZM alone could be developed. An
important aspect of the conjugated molecules is that due to the
presence of the polypeptide they are not expected to penetrate cell
walls, effectively excluding in vivo competition from all
intracellular carbonic anhydrases. Here we report on the
development of 4-C10L17-AZM, a polypeptide conjugate based
on the small organic inhibitor AZM and its binding interactions
with the carbonic anhydrases HCA II and HCA IX. It emerges as
a candidate molecule for the specific targeting of HCA IX and
could be used to deliver cargoes in a targeted way to HCA IX-
positive cancer cells detection, staging, imaging, and treatment.
1
13
1
was carried out by NMR spectroscopy ( H NMR, C NMR)
and mass spectrometry. The obtained active ester 1 was
determined to be pure and conjugated to the polypeptides without
further purification.
2
2
2
. Results and Discussion
.1. Design and synthesis
.1.1. Overall strategy
The overall development strategy included the design and
synthesis of an AZM derivative by conjugating it to a set of
polypeptides (Figure 1) followed by evaluation of the affinity for
HCA II by surface plasmon resonance (SPR) measurements. The
polypeptide conjugate with the highest affinity for HCA II was
also used to study interactions with three constructs designed to
mimic HCA IX. HCA II is readily available from commercial
sources whereas the full-length native membrane protein HCA IX
is very difficult to express. The construct HCA IX mimic is based
on soluble HCA II and has been mutated in the active site to
2
6
exactly match HCA IX. The construct HCA IX SV (SV for
surface variant), comprises the catalytic domain of HCA IX (from
amino acid 120 to 414) with 6 mutations on its surface to enhance
27
its crystallizability. The full extracellular domain of HCA IX
was expressed using the BEVS with the construct described by
16
Alterio, V. et al. For a detailed description of the expression
constructs and proteins used in this study, see the Experimental
part.
2
.1.2. Small molecule design
The AZM structure was modified to allow covalent attachment
to the side chain of a Lys residue via an aliphatic linker designed
not to interfere with the interactions between the AZM moiety
and the HCA active site. The design was based on the crystal
structure of the HCA II-AZM complex which is very similar to
2
5,16
Scheme 1. Synthetic route to active ester 1. Reagents and Conditions: a. 6
M HCl, reflux, 2 hours; b.methyl 10-chloro-10oxadecanoate, pyridine,
acetonitrile, 2 hours; c. 5% NaOH, 30 minutes; d.pentafluorophenol, EDCI,
pyridine, over-night.
that of the HCA IX-AZM complex.
The depth of the active
site pockets and the orientation of the bound AZM are very
similar for the two enzymes (Figure 2). The sulfonamide moiety
is the key to the design as the sulfonamide amino group
coordinates to the active site Zn ion whereas the acetyl group of
the AZM molecule is directed away from the pocket. In the
designed AZM derivative, the acetyl group was replaced by an
eight carbon bifunctional aliphatic linker with an active ester end
group to form 1 (Figure 1) which allowed convenient conjugation
to lysyl side chains in DMSO solution. The size of the linker was
based on a previous report on polypeptide conjugates developed
to bind HCA II but the synthesis strategy allowed for convenient
modification if the outcome of the affinity evaluation indicated
2
. 1.4. Polypeptide design
The polypeptide design has been described previously. In
1
short, the polypeptides were designed to fold into helix-loop-helix
motifs that dimerise to form four-helix bundles. At µM
concentrations in aqueous solution they were shown previously
by spectroscopic methods and analytical ultracentrifugation to
dimerize, whereas at nM concentrations the dimers dissociate to
3,29
form unordered monomers. The helical segments were
amphiphilic with four hydrophobic residues in each helix, and
solubility was ensured by the presence of charged and polar
groups (Figure 3). The polypeptide sequences differ only in the
positions where the small organic molecule is attached and with
regards to the number and nature of charged residues. Four
different positions where the small organic molecule is
2
8
that the linker size was not optimal.
2
.1.3. Small molecule synthesis
The synthesis starting material was AZM which is
commercially available. In step a (Scheme 1) the acetyl group
was removed by strong acid under reflux. After completion of the

Peptide code
Sequence
Global charge
2-C15L8-AZM
2-C10L17-AZM
2-C25L22-AZM
2-C37L34-AZM
3-C15L8-AZM
3-C10L17-AZM
3-C25L22-AZM
3-C37L34-AZM
4-C15L8-AZM
4-C10L17-AZM
4-C25L22-AZM
4-C37L34-AZM
Ac-NEADLEAKIRHLAEKLAARGPVDCAQLAEQLARAFEAFARKG-OH
Ac-NAADLEAAIKHLAEALKARGPVDCAQLAEQLARAFEAFARKG-OH
Ac-NEADLEAAIRHLAEALAARGPKDCKQLAEQLARAFEAFARKG-OH
Ac-NAADLEAAIRHLAERLAARGPVDCAQLAEQLAAKFEKFARKG-OH
-4
-4
-4
-3
-1
-1
-1
-1
+2
+2
+2
+2
Ac-NAADJEAKIRHLAEKJAARGPVDCAQJAEQLARRFEAFARKG-NH
Ac-NAADJEARIKHLAERJKARGPVDCAQJAEQLARAFEAFARKG-NH
Ac-NAADJEAAIRHLAERJAARGPKDCKQJAEQLARAFEAFARKG-NH
Ac-NAADJEAAIRHLAERJAARGPVDCAQJAEQLARKFEKFARKG-NH
2
2
2
2
Ac-NAADJEAKIRHLREKJAARGPRDCAQJAEQLARRFERFARKG-NH
Ac-NAADJEARIKHLRERJKARGPRDCAQJAEQLARAFERFARKG-NH
Ac-NAADJEARIRHLRERJAARGPKDCKQJAEQLARAFERFARKG-NH
Ac-NAADJEARIRHLRERJAARGPRDCAQJAEQLARKFEKFARKG-NH
2
2
2
2
Table 1. Sequences of polypeptide conjugates. AZM was covalently liked to the side chains of Lys residues shown in bold. Coumarin was coupled to the side
chains of Lys residues are shown as underscored. Cys-24 was Acm protected, Lys-41 was TFA protected and the N-terminal was acetylated in all sequences to
prevent conjugation to unprotected amino groups other than that designed for conjugation. J is norleucine. The suspected helical regions are residues 1-19 and 25-
4
2. GPVD is a commonly occurring turn sequence. Orthogonal protection groups were introduced to allow for the site-selective introduction of fluorophores,
small molecule ligands or immobilization on solid support. Lys residues to be functionalized on resin were protected by the Alloc group, and selectively removed
by Pd. Cys-24 was protected by the acetamidomethyl (Acm) group and deprotected by treatment with silver triflate (AgTfa). Lys-41 was protected by a
trifluoroacetyl group that is selectively removed by 90% aqueous piperidine.
2
. 1.6. Polypeptide conjugation
R15
R8
A30
The active ester 1 was not conjugated to the full set of
H11
D4
polypeptides. Instead twelve out of the sixteen peptides with
charges ranging from -4 to +2 were selected (Table 1) as they
were assumed to be more likely to bind to HCA II due to the
negative charge of the protein at neutral pH. Previous experience
has shown that few “hits” were obtained from the polypeptides
with total charge of -7. The active ester was reacted with purified
peptides in DMSO and the crude products were precipitated by
cold diethyl ether, lyophilized and purified by reversed-phase
HPLC. The fractions containing the desired product were
identified by MALDI-TOF MS and collected and lyophilized to
afford pure functionalized polypeptide conjugates.
RF 3 57
g
A32
AR 34 90
L12
J5
L31
g
c
FK 34 81
d
c
d
E14 A7
f
a
f EE 3 66 E29
b
a
A2
K41
b
A3
e
e
I9
A34
J27
R40
R33
K10
E6
R13
FR 33 57
Q30
J16
Figure 3. Schematic helical wheel representation of the suspected helical
segments in 4-C10L17-AZM, residues 2-16 (left) and residues 27-41 (right).
K10 is site of incorporation of coumarin, K41 is TFA protected. J is
norleucine.
2.2. Evaluation of affinities by SPR
2. 2.1. Measurement of affinities to HCA II with
protein immobilized
incorporated in combination with four different charges that vary
from -7 to +2 in steps of three, gave a total of sixteen sequences.
The polypeptide conjugates bind target protein considerably more
tightly and with increased selectivity, in comparison to the small
molecules alone, although the polypeptides by themselves do not
bind to a measurable degree under the experimental conditions.
Although there is some propensity for helix formation, the
flexibility of the polypeptide structure allows it to adapt to a
variety of protein surfaces and thus to enhance affinities for a
Evaluations of affinities for HCA II of the four polypeptide
conjugates with positive charge were investigated by SPR (SI
Figure S3). HCA II was immobilized on a CM-5 chip using a
standard amide coupling protocol, and each polypeptide conjugate
was included in the running buffer in the concentration range of
0
.24 to 62.5 nM. The ranking of affinities was based on the
maximum amount of uptake during association and the rate of
dissociation. All four polypeptide conjugates showed strong
binding to HCA II, as expected due to the affinity of the AZM
ligand, but based on the two criteria of off-rate and amount of
bound conjugate 4-C10L17-AZM and 4-C37L34-AZM appeared
to bind more strongly. The SPR analysis demonstrated that in
comparison with small molecule ligands the dominant effect of
polypeptide conjugation was to reduce the off-rate. The on-rates
were also reduced but only to a lesser extent. All four conjugates
2
-5,30-32
large number of proteins.
2
.1.5. Polypeptide synthesis
The 42-residue polypeptides were synthesized on the solid
phase using an automated peptide synthesizer and standard Fmoc
protocols. Before cleavage of the peptide from the resin, the N-
terminal was acetylated by subjecting it to an acetic anhydride
3
3
cocktail for 40 minutes ((CH
3
CH
2
CO)
2
O:DIPEA:NMP 10:5:85%
showed slower dissociation than the small molecule AZM itself.
(
v/v/v)). After acetylation, the alloc protecting group on the Lys
D
The K of 4-C10L17-AZM was determined to be 4 nM compared
side chain was selectively removed by Pd(0), followed by a
HCTU promoted coumarin incorporation procedure. All other
protecting groups were removed when the peptide was cleaved
from resin in TFA solution. The crude peptides were purified by
RP-HPLC and identified by MALDI-TOF mass spectrometry.
After cleavage, only one free Lys side chain remained in the
sequence available for conjugation to the active ester 1.
to 38 nM that was measured for AZM alone, an enhancement by a
factor of 10 (Figure 4; Figure 5). For affinity analysis based on
binding kinetics, BIAevaluation 4.1 (GE Healthcare Biosciences)
Rmax ×[L]describing a 1:1
was used. An equation, [RL]=
K +[L]
D
ꢀ

-
1
-1
-1
Polypeptide conjugate
k
a
/M s
SE(ka)
k
d
/s
SE(kd)
D
K /M
2
2
2
2
3
3
3
3
4
4
-C15L8-AZM
-C10L17-AZM
-C25L22-AZM
-C37L34-AZM
-C15L8-AZM
-C10L17-AZM
-C25L22-AZM
-C37L34-AZM
-C10L17-AZM
-C37L34-AZM
1.03E+05
1.19E+05
5.66E+04
1.37E+05
1.57E+05
1.11E+05
8.11E+04
1.84E+05
1.33E+05
1.08E+05
6.36E+02
1.52E+03
2.29E+03
1.15E+03
1.32E+03
6.83E+02
1.19E+03
7.48E+02
1.02E+03
3.58E+02
3.57E-03
3.61E-03
4.37E-03
3.44E-03
3.43E-03
4.09E-03
4.73E-03
1.07E-03
7.62E-04
1.94E-03
6.67E-06
1.58E-05
2.53E-05
1.17E-05
1.31E-05
7.35E-06
1.41E-05
6.25E-06
8.87E-06
3.09E-06
3.46E-08
3.03E-08
7.72E-08
2.52E-08
2.18E-08
3.70E-08
5.83E-08
5.84E-09
5.75E-09
1.79E-08
Table 2. Thermodynamic parameters describing interactions between polypeptide conjugates and HCA II measured at concentrations of 1nM, 10nM and
00nM. Binding kinetics were fitted using a 1:1 binding model available in BIAevaluation software. For sensorgrams, see SI, Figure S4.
1
binding model was applied to the experimental results where Rmax
is the total binding capacity, [L] is the concentration of the free 4-
C10L17-AZM, and [RL] is concentration of the complex. The K
of the AZM-HCA II complex was estimated from a competition
experiment where 4-C10L17-AZM was immobilized on the chip
and the running buffer contained preincubated AZM and HCA II.
binding. A low level of immobilization results in good
homogeneity of the recognition elements, and provides improved
quality of measured kinetic binding parameters. In contrast to
immobilized proteins, synthetic polypeptide conjugates are more
robust, do not denature and are more stable during long term
storage, apart from sustaining harsher conditions for chip
regeneration.
D
The K
D
of the HCA II-AZM complex was estimated from the
c , where c is the total concentration of
1.0
equation IC = K +
50
D
2
ꢀ
0
HCA II ([HCA II] = 50 nM) and IC50 is the concentration of
AZM needed to inhibit complexation of HCA II by 4-C10L17-
3
4
AZM to 50%. IC50 was estimated from the plot in Figure 4 to be
3 nM and the affinity of AZM for HCA II was thus estimated to
be 38 nM, in good agreement with the literature value of 20 nM
0.5
6
33
measured by SPR.
2
00
200
0.0
8
9
2
3
4
5
6
7
8
9
2
3
4
5
6
7
8
9
2
3
4
5
-
8
-7
10
-6
10
1
0
100
100
Figure 5. SPR biosensor analysis of HCA II – 4–C10L17-AZM complex
formation and its inhibition by AZM. An equation based on the competing
equilibria was fitted to the experimental data to give the dissociation constant
of 4 nM of the HCA II – 4–C10L17-AZM complex. 4-C10L17-AZM was
immobilized on chip and 50 nM HCA II was incubated with a series
concentration of AZM in HBS-P running buffer at pH 7.4 prior to injection.
Maximum response values are plotted versus AZM concentration. RUs are
normalized to a level of 1.0 for 50 nM HCA II.
0
4
00
500
600
700
800
10
20
30
40
-9
50x10
Time (s)
Conc.(M)
Figure 4. SPR biosensor analysis of the interactions between immobilized
HCA II and 4-C10L17-AZM in HBS-P buffer, at pH 7.4. The concentration of
4
-C10L17-AZM was 0.68, 2, 6.17, 18.5, 55.5 nM. The best fit of an equation
describing a 1:1 model to the maximum response as a function of the
concentration of polypeptide conjugates gave a K of 4 nM.
D
HCA II was injected as part of the running buffer at three
different concentrations, 1, 10 and 100nM and the sensorgrams
were evaluated as described above (Table 2). The parameters
measured for 4-C10L17-AZM and 4-C37L34-AZM were in good
agreement with those measured with immobilized HCA II
confirming the validity of our experimental setup. The conjugate
2.2.2. Measurements of affinities to HCA II with
peptides immobilized
To prevent unwanted degradation of immobilized HCA II
during the long term storage required for the SPR analysis of a
large set of peptide conjugates and to verify the experimental
setup the system was turned around. The polypeptide conjugates
4
-C10L17-AZM showed the slowest off-rate but all ten measured
(
Table 1) that have not previously tested as described in 2.2.1.
off-rates were within a factor of ten. This is in agreement with the
binder concept where the polypeptides enhance to a varying
extent the affinity of the AZM derivative. The dissociation
above with charges of -4 and -1 were immobilized on a CM-5
chip and, in addition, 4-C10L17-AZM and 4-C37L34-AZM were
included for comparison with the reverse measurements where
HCA II was immobilized. Cys-24 of each polypeptide was
deprotected and the free thiol group reacted with the maleimide
moiety of a bifunctional linker. The free amino group of the
bifunctional linker was then reacted with the carboxylate groups
of the SPR chip via conventional amine coupling procedure. The
lower molecular weights of the polypeptides in comparison to that
of HCA II required a lower level of immobilization to detect
D
constant of 4-C10L17-AZM measured this way, K = 5.8 nM, was
very close to that measured with the reverse setup, 4 nM. The
measurements using polypeptides immobilized on the chip proved
to be an efficient way of ranking sets of protein recognition
elements, provided of course that the protein was readily available
in sufficient amounts, as was HCA II. The small differences
between the polypeptide conjugates are difficult to analyze but

affinities were affected by the position where AZM was
incorporated as well as by the polypeptide conjugate charge.
concentration of 4-C10L17-AZM was in the range of 0.04–12.5
nM.
2.2.3. Inhibition of 4-C10L17-AZM binding to HCA
II by AZM
While there was no reason to assume that the best binding
polypeptide conjugate from the HCA II measurements should also
be the one that binds the most strongly to HCA IX, 4-C10L17-
AZM was selected for a preliminary evaluation of polypeptide
conjugate -HCA IX interactions.
In order to evaluate whether the interactions between 4-
C10L17-AZM and HCA II were guided exclusively by the
interaction between the AZM ligand and the protein, or if there
were nonspecific interactions between the polypeptide and HCA
II a competition experiment was set up as described above (Figure
4
-C10L17-AZM
AZM
HCA IX
0.09±0.014 nM
1±0.7 nM
3±0.6 nM
3±1.0 nM
0.6±0.3 nM
38 nM
5
). 4-C10L17-AZM was immobilized on the CM-5 chip and the
competition between AZM and 4-C10L17-AZM for HCA II was
measured as a function of AZM concentration (SI, Figure S5).
The formation of the HCA II - 4-C10L17-AZM complex was
reduced with increased AZM concentration, and at µM
concentration of ΑΖΜ complete inhibition was observed,
providing strong evidence that both 4-C10L17-AZM and AZM
bound to the same site. The results of the inhibition assay
demonstrated that 4-C10L17-AZM did not bind to HCA II
nonspecifically, and that improved binding was obtained from
peptide and/or aliphatic linker interactions with the protein.
HCA IX SV
HCA IX mimic
HCA II
1±0.3 nM
4±1.1 nM
Table 3. Dissociation constants measured by SPR analysis of the
complexes formed from 4-C10L17-AZM and Carbonic Anhydrases.
Surprisingly, the dissociation constant measured for 4-
C10L17-AZM and HCA IX was 90 pM, a factor of 33 lower than
D
that of AZM (Figure 7). AZM bound with a K of 3 nM (Table 3)
2.2.4. Measurements of affinities for HCA IX
variants
in good agreement with the previously reported values in the
35-37
range 6-20 nM.
The affinity enhancements observed for
polypeptide conjugates in comparison to AZM alone are due to
interactions between polypeptides and the target proteins.
HCA II and HCA IX mimic (the mimic is based on HCA II
with changed active site to match native HCA IX) differ by 7
residues in the active site, and share the same surface. Both differ
significantly from native HCA IX and HCA IX SV in their
surface charges (Figure 6). We observe effectively no difference
between HCA IX mimic and HCA IX SV, and only a small
improvement compared to HCA II (Table 3). This could be
attributed to a slightly enhanced affinity of the conjugated
compound towards the HCA IX active site architecture. However,
there is a dramatic improvement towards the native HCA IX
protein. This strongly suggests that the increased affinity of the
conjugate 4-C10L17-ACA towards native HCA IX is due to an
interaction with the PG domain. An alternative explanation is that
the surface mutations introduced in the HCA IX SV protein
interferes with peptides interactions. If the targeting is specific for
the PG domain, this is very promising and suggests that tailored
reagents for HCA IX detection or inhibition can be developed as
the PG domain is unique to HCA IX compared to the other
HCAs. It has been suggested that the PG domain enhances the
catalytic activity of CA IX and plays a role in CA IX mediated
Figure 6. Surface representation of HCA IX constructs used in this study
with charge indicated in blue (positive) and red (negative). Please note that the
surface of HCA II is the same as shown for HCA IX mimic here. HCA IX
mimic as described by Genis et al. is HCA II with the introduction of 7 point
mutations in the active site of HCA II to create a mimic of the active site of
2
6
HCA IX. For more details on the different HCA IX constructs please see SI,
Figure S1 and S2. Figure was generated using PyMOL (The PyMOL
Molecular Graphics System, Version 1.7, Schrödinger, LLC)
In order to determine thermodynamic parameters for
interactions between AZM and the extracellular domain of HCA
IX, three variants were produced (SI, Figure S1; Figure S2). Full
length native HCA IX contains the CA catalytic and proteoglycan
38
(
PG) domain on the extracellular side as well as the membrane
binding to the extracellular matrix.
spanning helix and intracellular domain. The PG domain is
unique to HCA IX compared to other HCAs. We expressed and
purified extracellular HCA IX, (CA + PG domains without the
membrane spanning helix and intracellular domains) from insect
cells. In addition, a mutant of HCA II with its active site identical
to that of HCA IX (HCA IX mimic) and the catalytic domain only
of HCA IX (with 6 additional surface mutations, HCA IX SV)
were both expressed in E.coli. A comparison of the surface charge
of the catalytic domains of all three HCA IX variants used in this
study is shown in Figure 6. In this orientation, there are clearly
many variable regions in terms of charge on the surface of HCAs.
HCA IX appears to have a more positive area immediately
outside of the active site compared to HCA IX mimic (i.e. surface
of HCA II).
The affinities of AZM for all three variants were almost
identical, providing strong evidence that the protein structures
with regards to interactions in the active site were not affected by
removal of PG domain. It may be that polypeptide conjugates
other than 4-C10L17-AZM bind HCA IX with higher affinity but
the demonstration that a 33-fold affinity enhancement could be
achieved for a small organic molecule starting with low nM
affinity demonstrated that very high affinities are within reach
D
using polypeptide conjugation technologies. The K of 90 pM is
sufficient for a variety of needs in biotechnology where it would
take considerable efforts to reach similar performance using
traditional strategies. The outcome of the investigation also
demonstrated that it is possible to modify molecules structurally
without perturbing the finely tuned interactions required for low
nM affinities. In comparison to small organic molecules
polypeptide conjugates are likely to demonstrate very different
biodistributions, not penetrating to any great degree cell
membranes, with the exception to those where the polypeptide
Using a standard amide coupling protocol, the three protein
constructs were immobilized on a CM-5 chip and their
interactions with 4-C10L17-AZM evaluated by SPR analysis
(
Figure 7; SI, Figure S6; Figure S7). The concentration of AZM
in the running buffer was in the range of 0.78–100 nM, and the

conjugate binds very strongly. Polypeptide conjugates may
therefore be of interest in medical imaging of tumors.
All reagents were purchased from commercial sources and
were used without further purification unless otherwise noted.
Evaporations in small molecule synthesis were performed at
reduced pressure rotatory evaporator. Crude peptides and peptide
conjugates were performed by HPLC (Varian Prostar,
semipreparative column: pore size 100 Å, particle size 10 µm,
dimensions 250x21.2 mm, C-8 semi-preparative column).
Gradient and method: acetonitrile/water (30%-60%) in 30
minutes, with 0.1% TFA added, UV detection at 210 nm
wavelength. HPLC fractions were analyzed off-line by MALDI-
TOF MS using α-cyano-4- hydroxycinnamic acid as matrix. In
small organic synthesis, ESI-MS was recorded using an Agilent
1
1
0
0
0
.2
.0
.8
.6
.4
1
0
1
-
200
300
0
100
200
Conc. (M)
-9
3
00x10
Time (s)
1
5
1
1
5
0
5
0
10
1
100/Waters Micromass ZQ instrument using a 3.0x50 mm C18
column and a gradient elution with 5:95% acetonitrile-water (with
.1% formic acid) to 95:5% acetonitrile-water (with 0.1% formic
5
0
2
00
400
600
800
1000
0
2
4
6
8
10
-9
1
2x10
acid). The mass detector was operated in alternating positive and
negative ion modes. NMR spectra were recorded at 400 MHz
Varian Unity spectrometer and 298K using DMSO-d6 as the
solvent (internal DMSO-H6, δH 2.50, δC 39.52 ppm). For
Time (s)
Conc.(M)
Figure 7. SPR sensorgrams showing interactions between immobilized
HCA IX and AZM (top left), and between immobilized HCA IX and 4-
C10L17-AZM (bottom left) in HBS-P buffer with 3% DMSO, at pH 7.4. For
AZM binding assay, a concentration series (1.3, 4.1 12.3, 37, 111, 333 nM)
was used. For 4-C10L17-AZM binding assay, a concentration series (48.7,
1
13
Mass spectra, H and C NMR spectra of intermediates and
product, see SI, Figure S8.
4
. 1.1. Synthesis of 1
-amino-1,3,4-thiadiazole-2-sulfonamide (2). Acetazolamide
4.44 g, 20 mmol) was suspended in 6 M HCl and refluxed for 2
9
7.5, 195, 390, 780 pM, 1.56, 3.12 6.25, 12.5nM) was used. Fits of an
39
5
equation describing a 1:1 equilibrium to the maximum responses as a function
of the concentration of binder molecules are shown next to each sensorgram,
(
hours. TLC (EA/PE 1/9). The reaction mixture was neutralized
with concentrate NaOH solution. The precipitate was filtered and
recrystallized from ethanol. Yield 2.20 g (51%). HR-MS (m/z):
D
giving a K of 3 nM for AZM and 90 pM for 4-C10L17-AZM.
3
. Conclusion
+
13
[
M+H] (calculated for C
2 5 4 2 2
H N O S ): 180.9858 (180.9854). C
A polypeptide conjugate, 4-C10L17-AZM, has been developed NMR δ 171.58, 158.01.
that recognizes HCA II and HCA IX with affinities that are larger
than that of the small molecule ligand AZM alone by factors of 10
and 33, respectively. 4-C10L17-AZM binds HCA IX with a
dissociation constant of 90 pM and a comparison with other
constructs strongly suggests that the difference may be due to
interactions with the PG domain. In addition, the results from
experiments with the HCA IX surface variant protein support the
concept that the conjugated peptides target areas remote from the
active site and in a specific way. This is an exciting result as the
PG domain is unique to HCA IX compared to the other HCAs and
provides a specific target. The results also suggest that it is
possible to target the cancer marker HCA IX in bioanalytical and
biomedical applications using a construct that is unlikely to
penetrate cell membranes and is much smaller than a full
monoclonal antibody.
Methyl 5-(9’-carboxyldecanylamino)-1,3,4-thiadiazole-2-
sulfonamide (3). 2 (650 mg, 3 mmol) was dissolved in acetonitrile
15 mL) and pyridine (1.5 mL) was added. Methyl 10-chloro-10-
oxadecanoate (669 µL, 3 mmol) was added dropwise and the
mixture stirred for 2 hour at room temperature. TLC
(
(
DCM/MeOH 10/1). The solvent was evaporated, 20 mL 2 M
HCl solution was poured to the reaction residue. The precipitate
was collected by filtration and was recrystallized from ethanol.
+
Yield 711 mg (63%). HR-MS (m/z) [M+H] (calculated for
1
C
13
H
23
N
4
O
5
S
2
): 379.1120 (379.1110). H NMR δ 12.95 (s, 1H),
8
2
.29 (s, 2H), 3.54 (s, 3H), 2.52 – 2.44 (m, 2H), 2.25 (t, J = 7.3 Hz,
H), 1.64 – 1.52 (m, 2H), 1.52 – 1.41 (m, 2H), 1.29-1.16 (m, 8H).
C NMR δ 173.79, 172.70, 164.68, 161.48, 51.59, 35.20, 33.67,
8.92, 28.90, 28.82, 28.79, 24.83, 24.78.
1
3
2
5
-(9’-methoxylcarbonyldecanylamino)-1,3,4-thiadiazole-2-
sulfonamide (4). 3 (379 mg, 1 mmol) was dissolved in NaOH (5
) solution and stirrred for 30 minutes at room temperature. TLC
DCM/MeOH 10/1). The reaction mixture was acidified with
4
. Experimental part
%
(
Abbreviations:
AZM acetazolamide,
concentrated HCl to acidic pH around 2, the precipitate was
collected by filtration and was recrystallized from ethanol. Yield
DCM dichloromethane,
+
1
C
85 mg (51%). HR-MS (m/z): [M+H] (calculated for
1
DIPEA diisopropylethylamine,
EDCI N-Ethyl-N′-(3-dimethylaminopropyl)carbodiimide
12
H
21
N
4
O
5
S
2
): 365.0964 (365.0953). H NMR δ 12.96 (s, 1H),
8
1
.29 (s, 2H), 2.49 (t, J = 7.3 Hz, 2H), 2.15 (t, J = 7.4 Hz, 2H),
.64 – 1.52 (m, 2H), 1.49 – 1.39 (m, 2H), 1.29-1.19 (m, 8H).
1
3
C
hydrochloride,
NMR δ 174.92, 172.74, 164.63, 161.54, 35.24, 34.07, 28.99,
28.94, 28.84, 24.90, 24.81, (one aliphatic carbon missing).
EA ethyl acetate,
MQ mili-Q water,
NMP N-methylpyrrolidone,
PE pentane,
5-(9’-pentafluorophenoxylcarbonyldecanylamino)-1,3,4-
thiadiazole-2-sulfonamide (1). 4 (16 mg, 0.04 mmol) was
dissolved in pyridine 9 mL, then to the solution,
pentafluorophenol (14.7 mg, 0.08 mmol) in 1 mL pyridine was
added, followed by EDCI (16.9 mg, 0.08 mmol). The reaction
mixture was stirred overnight at room temperature. The solvent
was concentrated, then 10 mL EA was added to dissolve the
residue. EA solution was washed by water 10 mL 2 times. The
combined EA phase was dried with anhydrous sodium sulfate and
TFA trifluoroacetic acid.
4
.1. Chemistry

was concentrated to give 1. Afforded 2 was then subjected to a
high vacuum overnight before usage. HR-MS (m/z): [M+H]
stock solution (10mg/mL) to a solution of the polypeptide in
DMSO (1 mM) in the presence of 5 equivalents of DIPEA. The
reaction was monitored by analytical HPLC, and the fractions
were analyzed by MALDI-TOF MS. The reaction mixture was
quenched with water, lyophilized and then purified by preparative
RP-HPLC. Collected fractions were characterized by MALDI-
TOF MS and the appropriate fraction was lyophilized (in yields
30-50%).
+
(
calculated for C18H N O S F ): 531.0801 (531.0795). The crude
20 4 5 2 5
1
active ester 2 was used directly in peptide conjugation step. H
NMR δ 8.32 (s, 2H), 2.77 (t, J = 7.3 Hz, 2H), 2.52 (d, J = 7.3 Hz,
1
3
2
1
2
H), 1.70 – 1.56 (m, 4H), 1.36 – 1.24 (m, 8H). C NMR δ
72.70, 170.00, 164.65, 161.51, 35.21, 32.89, 28.91, 28.77, 28.49,
4.76, 24.66.
4
4
.1.2. Peptide synthesis and conjugation
.1.2. 1. Peptide synthesis
4. 1.3. Preparation of peptide conjugates for chip
coupling
4
. 1.3.1. Acm deprotection
Chemical reagents were purchased from commercial sources
and were used without further purification unless noticed. The
polypeptides were synthesized on a peptide synthesizer (Prelude,
Protein Technologies Inc.) using standard Fmoc chemistry with
HCTU (ChemPep ) and DIPEA (Sigma Aldrich) as activating
reagents. Fmoc deprotection of the amino terminal was performed
using 20% piperidine in NMP. The synthesis was performed in a
The peptide (1.7 mg) was dissolved in TFA (500 µL) and
anisol (10 µL) was added to the peptide solution. The solution
o
was stirred and cooled to 0 C. Silver triflate (AgOTf, 13 mg) was
®
added to the cold peptide solution. The solution was stirred at 0
C for 1 hour and the temperature was allowed to come to room
o
temperature and the stirring was continued overnight. The silver
salt of the peptide was precipitated and centrifuged. The
supernatant was discarded and the residue was dissolved in 35%
acetic acid (700 µL). DTT (5mg) was added to the peptide
solution, after that the solution was stirred for 2 hours. The
reaction mixture was centrifuged and the peptide was purified by
RP-HPLC. Collected fractions were characterized by MALDI-
TOF MS and the appropriate fraction was lyophilized (in 50%
average yield).
1
00 µmol scale using rink amide resin (0.44 mmol/g) or Wang
resin preloaded with desired amino acid (0.45 mmol/g, Sigma
Aldrich). To the resin, 5 folds of amino acid, 5 folds of HCTU
(
0.45 M in DMF) and 10 folds of DIPEA (2 M in NMP) were
sequentially added, the coupling was performed twice and 20
minutes for each. The side chains of the amino acids (Iris Biotech
GmbH) were protected by either base-stable groups: tert-butyl
ester (Asp and Glu), trt (His, Asn, and Gln), 2,2,4,6,7-
pentamethyldihydrobenzofuran-5-sulfonyl (Arg) and
acetamidomethyl (Cys) or by Tfa (Lys) which is stable during
peptide synthesis, cleavage and conjugation.
4
. 1.3.2. Introducing malaimide linker
Peptide with free SH (1.5 mg) was dissolved in 1mL DMSO.
Subsequently, 30 µL 1-(2-Amino-ethyl)-pyrrole-2,5-dione
hydrochloride (18mg/mL) was added to the solution. The reaction
was flushed with N and stirred for 2 hours. After lyophilization
2
of reaction mixture, the crude product was purified by RP-HPLC.
Collected fractions were characterized by MALDI-TOF MS and
the appropriate fraction was lyophilized (in 90% average yield).
For the site-selective incorporation of a fluorescent probe 7-
methoxycoumarin-3-carboxylic acid and ligands, the side chains
of lysine residues were N-protected by an Alloc group, which can
be selectively deprotected by treatment with tetrakis-
(
triphenylphosphine) palladium (0) (Pd(PPh ) ), or by a Boc
3 4
group, which can be deprotected by TFA. After complete
machine synthesis, the resin was rinsed with DMF and DCM and
dried in a desiccator under vacuum.
4
.2. Screening and affinity measurements with SPR
General. All experiments were performed with a Biacore 2000
Fmoc group on N-terminal was deprotected by 20%
piperidine. The N-terminal of the polypeptide was capped
manually with a mixture of acetic anhydride, DIPEA, and NMP
instrument at 298 K. HCA II was purchased from SigmaAldrich.
Amide coupling kit, HBS-P buffer, coupling buffer and CM5
chips were purchased from GE Healthcare Biacore. HBS-P was
used as running buffer in the coupling procedure; HBS-
P+3%(v/v) DMSO was prepared and used as running buffer in
kinetic assay unless specially mentioned. In each assay, ligand
was immobilized on cell 2, and cell 1 with deactivated dextran
was used as reference.
(
v/v/v 10:5:85) for 20 minutes. After washing, alloc- on lysine
was deprotected with a solution of (Pd(PPh (30 mM in 10%
acetic acid, 85% CHCl and 5% N-methyl morpholine, 10 mL)
and agitated under N for 3 h. The resin was filtered, washed
3 3
)
3
2
consecutively with 0.5% DIPEA in DMF (3 x 10 mL), 0.5%
sodium dithiocarbamate in DMF (3 x 10 mL), DMF (3 x 10 mL),
DCM (3 x 10 mL) and dried under vacuum until dryness. The
fluorescence probe 7-methoxycourmarin-3-carboxylic acid was
coupled to the amino group of lysine side chain was performed in
DMF (10 mL) for 2 hours at room temperature with two
equivalents of 7-methoxycourmain-3-carboxylic acid, two
equivalents of HCTU and four equivalents of DIPEA. The resin
was then washed with DMF and DCM, and then the resin was
desiccated. The peptide was cleaved from the resin by 5 mL of
4. 2.1. Peptide conjugates screening and 4-C10L17-
AZM affinity measurement on HCA II immobilized
chip
To immobilize HCA II on chip, the protein was dissolved in 10
mM sodium acetate buffer, pH 5.5, to concentration 200 µg/mL.
Immobilization procedure was performed by Application Wizard.
The targeting immobilization level was set to 1500 RU, the
resulting immobilization level was 997 RU. In reference cell,
activation was performed for 7 minutes then 7 minutes
deactivation by ethanolamine pH 8.0.
9
2
5% TFA, 2.5% TRIS, 2.5% H O, the reaction time was 1 hour.
TFA solution was separated from resin by filtration. The crude
peptide obtained by cold ether precipitation in TFA solution was
then dissolved in water and lyophilized overnight. The crude
peptide was purified by RP-HPLC and characterized by MALDI-
TOF MS (positive mode).
In the assay, 4-C15L8-AZM, 4-C10L17-AZM, 4-C25L22-
AZM, and 4-C37L34-AZM were dissolved in DMSO (1.8
mg/mL) as stock solution. For each sequence, a concentration
series (0.24, 0.97, 3.9, 15.4, 62.5 nM) containing 3% DMSO were
injected for 3 minutes, following dissociation for 7 minutes. At
the end of each cycle, regeneration was performed by running
buffer clearance for 60 minutes.
4
.1.2. 2. Peptide conjugation
The ligands were conjugated to the free lysyl amino group of
the peptides by the addition of 2.5 equivalents of 1 in DMSO

4
-C10L17-AZM affinity measurement was performed on a
HCA IX containing the full extracellular domain (signal peptide,
PG domain and CA catalytic domain of HCA IX) with a 8 x His-
tag and Thrombin-cleavage site inserted between the signal
peptide and PG domain was synthesized by GeneArt
new protein coupled chip. By following the same procedure,
analysis was assayed in a concentration series (0.68, 2, 6.17, 18.5,
5.5 nM) containing 3% DMSO.
5
(
ThermoFisher) and subcloned into pVL1393 (BD Bioscience).
4
.2.2. Screening using peptide conjugates
Recombinant HCA IX using the BEVS was produced using the
using BaculoGold DNA (BD Bioscience) to generate the
recombinant baculovirus. HCA IX expression was done using Sf9
cells (Invitrogen) in SFX-Insect (HyClone GE Healthcare) cell
culture medium, infection at MOI 2 and harvesting of the cell
culture medium by centrifugation 4 d after infection. Purification
of HCA IX was done by immobilized metal affinity
immobilized on chip
For each sequence of modified peptides_with S -
malaimide(C )-NH , the peptide was dissolved in 10 mM maleic
acid (pH 6.0) to concentration 1 mg/mL. Immobilization was
performed by activation for 7 minutes, coupling for 10-20
minutes. The immobilizing level differed by each peptide (from
2
4
2
2
1
00 RU to 400 RU). In reference cell, the same procedure was
chromatography using HisTrap Excel (GE Healthcare) columns
equilibrated in buffer A (50 mM NaPO , 500 mM NaCl, pH8.0).
4
performed without peptide coupling. In the assay, HBS-P was
used as running buffer and the flow rate was set to 50 µL/min. A
concentration series of HCA II (1 nM, 10 nM, 100 nM) were
injected to peptide immobilized surface for 3 minutes,
dissociation for 7 minutes. The surface was regenerated by 30s
pulse of 10 mM NaOH followed with 30s pulse of 10 mM HCl.
Stabilization was left for 2 minutes.
Elution of bound protein was achieved by increasing
concentrations of imidazole up to 500 mM in buffer A.
Subsequent purification was done by size exclusion
TM
chromatography using Superdex 200 pg resin (GE Healthcare)
in 50 mM Tris/HCl pH 8.5. In comparison to the reports by Hilvo
et al. and Alterio et al. we did not cleave away the 8 x His-tag and
thereby we also did not observe degradation of HCA IX during
the purification procedure as reported by both e.g., that a band
migrating at the molecular mass of the sole CA domain was
4
.2.3. AZM inhibition assay
To the 4-C10L17-AZM immobilized chip, pre-incubated HCA
II 50 nM with a concentration series of AZM (0, 7.8, 15.6, 31.2,
2
2,16
observed.
6
2.5, 125, 250, 500, 1000, 5000 nM) was injected for 3 minutes
and allowed dissociation for 20 minutes. For inhibition study, the
responding RU from binding 1 in the association phase was
plotted against to the concentration of AZM.
4
. 3.2. HCA IX SV
The surface variant (SV) of HCA IX, was described previously
12
by Mahon et al. It is the catalytic domain of HCA IX that
4.2.4. 4-C10L17-AZM affinity measurement on HCA
IX, HCA IX SV and HCA IX mimic
contains 6 surface mutations that provide more solubility and
40
stability to the enzyme. The cDNA encoding HCA IX SV was
synthesized and purchased from GenScript. The HCA IX SV
coding region was cloned into pET-26b(+) (Novagen) using NdeI
Size exclusion chromatography buffer of HCA IX SV, HCA
IX mimic and HCA IX were exchanged into coupling buffer
/
(
XhoI. HCA IX SV was expressed in BL21 Star (DE3) cells
Invitrogen) in LB Broth, Miller (Difco) supplemented with 100
(
sodium acetate pH 5.0/5.5/5.5) by using PD-10 desalting
columns. The concentration of the protein was then calibrated to
µg/mL kanamycin. Cells were grown to an OD600 1.0, at which
point expression was induced with 1 mM isopropyl β-D-1-
thiogalactopyranoside for 4 h at 37 °C in the presence of 1 mM
2
00 µg/mL for the next step use.
Immobilization was performed by activation for 7 minutes
followed by coupling for 10 minutes. The same procedure was
repeated the reference cell, except of injection of protein. The
level of immobilization of HCA IX, HCA IX SV and HCA IX
corresponded to 2980, 1443 and 577 RUs respectively.
ZnSO
by centrifugation at 6000 x g, and disrupted by French press in
.2 M Na SO , 100 mM Tris/HCl pH 9.0. Cell lysate was clarified
by centrifugation at 14000 rpm for 1 h at 4 °C prior to purification
by affinity chromatography using a gravity-fed column with p-
amino-methyl-benzenesulfonamide resin (Sigma Aldrich).
Nonspecific bound proteins were washed off by repeated wash
4
. The cells were incubated for additional 4 hours, harvested
0
2
4
In the assay, 1 M acetazolamide and 500 µM 4-C10L17-AZM
in DMSO was used as stock solution. For AZM binding assay, a
concentration series (0.78, 1.56, 3.12 6.25, 12.5, 25, 50, 100 nM)
containing 3% DMSO was used for HCA IX SV, a concentration
series (0.78, 1.56, 3.12 6.25, 12.5, 25 nM) containing 3% DMSO
was used for HCA IX mimic and a concentration series (1.3, 4.1
steps with wash buffers (0.2 M Na
.0 and pH 9.0). HCA IX SV was eluted with 50 mM Tris/HCl
pH 7.8, 0.4 M NaN . The eluted protein was concentrated with
2 4
SO , 100 mM Tris/HCl at pH
7
3
Ultra centrifugation filtration devices using 10 kDa molecular
1
2.3, 37, 111, 333 nM) containing 3% DMSO was used for HCA
weight cut off (Amicon) and thereafter subjected to gel filtration
IX. Flow rate was set to 50 µL/min. Injection time: 1 minute.
Dissociation time: 3 minutes. For 4-C10L17-AZM binding assay,
a concentration series (48.7, 97.5, 195, 390, 780 pM, 1.56, 3.12
TM
size exclusion chromatography using Superdex 75 pg resin (GE
Healthcare) in 50 mM Tris pH 8.5.
6
.25, 12.5, 25, 50 nM) was used. Flow rate was set to 30 µL/min.
4
. 3.3. HCA IX mimic
Injection time: 1.5 minutes. Dissociation time: 10 minutes.
Additional 40 minutes was used for clearance.
The HCA IX mimic was first described by Genis et al. by
introducing 7 point mutations in the active site of human CA II to
2
6
create a mimic of the active site of human CA IX. The
expression plasmid for HCA IX mimic was a kind gift from Prof
R. McKenna, University of Florida, USA. HCA IX mimic was
expressed in E. coli BL21(DE3) cells as described by Pinard et
al.. The purification of HCA IX mimic was done as described
for HCA IX SV above.
4.3. Expression and purification of HCA IX, HCA IX SV and HCA
IX mimic Proteins.
Human Carbonic Anhydrase II was purchased from a
commercial supplier (Sigma-Aldrich, C6165) and used without
further purification.
23
4
.3.1. HCA IX (full extracellular part of human CA
IX)
HCA IX production using the baculovirus expression vector
system (BEVS) was modified after the methods described by
A schematic representation of the different HCA IX constructs
is given in Figure S1. The amino acid sequences of all
recombinant proteins described above can be found in Figure S2.
It also shows the amino acid sequence of HCA II in comparison.
2
2,16
24
Hilvo et al. and Alterio et al..
The expression construct for
Sequence alignments were made using Clustal Omega.

Acknowledgments
24. Moeker, J.; Mahon, B. P.; Bornaghi, L. F.; Vullo, D.; Supuran, C.
T.; McKenna, R.; Poulsen, S.-A. J. Med. Chem. 2014, 57, 8635–
8
645.
We thank the Swedish Research Council (L.B grant number
014-4057) Lund University, Royal Physiographic Society of
2
5. Fisher, S. Z.; Aggarwal, M.; Kovalevsky, A. Y.; Silverman, D. N.;
2
McKenna, R. J. Am. Chem. Soc.. 2012, 134, 14726–14729.
Lund and The Crafoord Foundation (W.K., K.K) for financial
support. ZF and WK would like to thank Prof. Robert McKenna
and Dr. Brian Mahon for providing materials (constructs).
26. Genis, C.; Sippel, K. H.; Case, N.; Cao, W.; Avvaru, B.
S.; Tartaglia, L. J.; Govindasamy, L.; Tu, C.; Agbandje-
McKenna, M.; Silverman, D. N.; Rosser, C. J.; McKenna,
R. Biochemistry 2009, 48, 1322–1331.
References
27. Mahon, B. P.; Bhatt, A.; Socorro, L.; Driscoll, J. M.; Okoh, C.;
Lomelino, C. L.; Mboge, M. Y.; Kurian, J. J.; Tu, C.; Agbandje-
McKenna, M.; Frost, S. C.; McKenna, R. Biochemistry 2016, 55,
1
2
.
.
Baltzer, L. Anal. Bioanal. Chem. 2011, 400, 1653–1664.
Yang, J.; Gustavsson, A.-L.; Haraldsson, M.; Karlsson, G.; Norberg,
T.; Baltzer, L. Org. Biomol. Chem. 2017, 24, 337–4654.
Tegler, L. T.; Nonglaton, G.; Büttner, F.; Caldwell, K.; Christopeit,
T.; Danielson, U. H.; Fromell, K.; Gossas, T.; Larsson, A.; Longati,
P.; Norberg, T.; Ramapanicker, R.; Rydberg, J.; Baltzer, L. Angew.
Chem. Int. Ed. 2011, 50, 1823–1827.
4
642–4653.
8. Enander, K.; Dolphin, G. T.; Baltzer, L. J. Am. Chem. Soc. 2004,
26, 4464– 4465.
9. Olofsson, S.; Johansson, G.; Baltzer, L.J. Am. Chem. Soc. 1995, 0,
047–2056.
0. Sun, X.; Yang, J.; Norberg, T.; Baltzer, L. J. Pept. Sci. 2012, 18,
31–739.
1. Ramapanicker, R.; Sun, X.; Viljanen, J.; Baltzer, L. Bioconjug.
Chem. 2013, 24, 17–25.
2. Ślósarczyk, A. T.; Baltzer, L. Org. Biomol. Chem. 2011, 9, 7697–
2
2
3
3
3
3
1
3
.
2
7
4
.
Tegler, L. T.; Fromell, K.; Jonsson, B.-H.; Viljanen, J.; Winander,
C.; Carlsson, J.; Baltzer, L. ChemBioChem 2011, 12, 559–566.
Balliu, A.; Baltzer, L. ChemBioChem. 2017, 14, 1.
Supuran, C. T. J. Enzyme Inhib. Med. Chem. 2012, 27, 759–772.
Davis, A. M.; Teague, S. J. Angew. Chem. Int. Ed. 1999, 38, 736-
5
6
7
.
.
.
7
704.
3. Rogez Florent, T.; Duhamel, L.; Goossens, L.; Six, P.; Drucbert, A.
S.; Depreux, P.; Danzé, P. M.; Landy, D.; Goossens, J. F.; Foulon,
C. J. Mol. Recogn. 2014, 27, 46–56.
7
49.
Bissantz, C.; Kuhn, B.; Stahl, M. J. Med. Chem. 2010, 53, 5061-
084.
8
9
.
.
5
3
3
3
4. GE Health. (2001). Biacore Concentration Analysis
Handbook. Retrieved
McKenna, R.; Supuran, C. T. In Carbonic Anhydrase: Mechanism,
Regulation, Links to Disease, and Industrial Applications. Subcell
Biochem. 2014. 75, 291–323.
from https://www.biacore.com/lifesciences/service/downloads/Hand
books/index.html?section=lifesciences&realsection=lifesciences.
5. Bruno, E.; Buemi, M. R.; Di Fiore, A.; De Luca, L.; Ferro, S.;
Angeli, A.; Cirilli, R.; Sadutto, D.; Alterio, V.; Monti, S. M.;
Supuran, C. T.; De Simone, G.; Gitto, R. J. Med. Chem. 2017, 60,
acs.jmedchem.7b00264–4326.
1
1
0. Kanfar, N.; Tanc, M.; Dumy, P.; Supuran, C. T.; Ulrich, S.; Winum,
J.-Y. Chem. Eur. J. 2017, 23, 6788–6794.
1. Jung, H. S.; Han, J.; Shi, H.; Koo, S.; Singh, H.; Kim, H.-J.; Sessler,
J. L.; Lee, J. Y.; Kim, J.-H.; Kim, J. S. J. Am. Chem. Soc. 2017, 139,
7
595–7602.
2. Gokhale N.H.; Bradford S.; Cowan, J. A. J. Am. Chem. Soc. 2008,
30, 2388–2389.
3. Silverman, D. N.; McKenna, R. Acc. Chem. Res. 2007, 40, 669–675.
4. Alterio, V.; Di Fiore, A.; D'Ambrosio, K.; Supuran, C. T. De
Simone, G. Chemical Rev. 2012, 112, 4421–4468.
6. Dudutienė, V.; Matulienė, J.; Smirnov, A.; Timm, D. D.; Zubrienė,
1
A.; Baranauskienė, L.; Morku
Michailovienė, V.; Juozapaitienė, V.; Mickevičiu
J.; Bakšytė, S.; Kasiliauskaitė, A.; Jachno, J.; Revuckienė, J.;
Kišonaitė, M.; Pilipuitytė, V.; Ivanauskaitė, E.; Milinavičiutė, G.;
̅
naitė, V.; Smirnovienė, J.;
1
̅
tė, A.; Kazokaitė,
1
1
̅
Smirnovas, V.; Petrikaitė, V.; Kairys, V.; Petrauskas, V.; Norvaišas,
P.; Lingė, D.; Gibieža, P.; Čapkauskaitė, E.; Zakšauskas, A.;
Kazlauskas, E.; Manakova, E.; Gražulis, S.; Ladbury, J. E.; Matulis,
D. J. Med. Chem. 2014, 57, 9435–9446.
1
1
5. Krishnamurthy, V. M.; Kaufman, G. K.; Urbach, A. R.; Gitlin, I.;
Gudiksen, K. L.; Weibel, D. B.; Whitesides, G. M. Chemical Rev.
2
008, 108, 946–1051.
6. Alterio, V.; Hilvo, M.; Di Fiore, A.; Supuran, C. T.; Pan, P.;
Parkkila, S.; Scaloni, A.; Pastorek, J.; Pastorekova, S.; Pedone, C.;
Scozzafava, A.; Monti, S. M.; De Simone, G. Proc. Natl. Acad. Sci.
U.S.A. 2009, 106, 16233–16238.
3
3
7. Wingo, T.; Tu, C.; Laipis, P. J.; Silverman, D. N. Biochem. Biophys.
Res. Commun. 2001, 288, 666–669.
8. Csaderova, L.; Debreova, M.; Radvak, P.; Stano, M.; Vrestiakova,
M.; Kopacek, J.; Pastorekova, S.; Svastova, E. Front. Physiol. 2013,
1
1
7. van Kuijk, S. J. A.; Yaromina, A.; Houben, R.; Niemans, R.;
Lambin, P.; Dubois, L. J. Front Oncol 2016, 6, 69.
4
, 271.
3
4
9. Sayed, N. S.; Bendary, E. R.; Ashry, S. M.; Kerdawy, M. M. Eur. J.
Med. Chem. 2011, 46, 3714–3720.
0. Teicher, B. A.; Liu, S. D.; Liu, J. T.; Holden, S. A.; Herman, T. S.
Anticancer Res 1993, 13, 1549–1556.
8. Lou, Y.; McDonald, P. C.; Oloumi, A.; Chia, S.; Ostlund, C.;
Ahmadi, A.; Kyle, A.; Keller, U. A. D.; Leung, S.; Huntsman, D.;
Clarke, B.; Sutherland, B. W.; Waterhouse, D.; Bally, M.;
Roskelley, C.; Overall, C. M.; Minchinton, A.; Pacchiano, F.; Carta,
F.; Scozzafava, A.; Touisni, N.; Winum, J.-Y.; Supuran, C. T.;
Dedhar, S. Cancer Res. 2011, 71 (9), 3364–3376..
1
2
9. Ward, C.; Meehan, J.; Mullen, P.; Supuran, C.; Dixon, J. M.;
Thomas, J. S.; Winum, J.-Y.; Lambin, P.; Dubois, L.; Pavathaneni,
N.-K.; Jarman, E. J.; Renshaw, L.; Um, I.; Kay, C.; Harrison, D. J.;
Kunkler, I. H.; Langdon, S. P. Oncotarget. 2015, 6, 24856–24870.
0. Parks, S. K.; Cormerais, Y.; Durivault, J.; Pouysségur, J.
Oncotarget. 2017, 8, 10225–10237.
Supplementary Material
Supplementary data associated with this article can be found.
2
2
1. Pastorek, J.; Pastorekova, S. Semin. Cancer Bio. 2015, 31, 52–64..
2. Granja, S.; Tavares-Valente, D.; Queir s, O. L.; Baltazar, F. T.
Seminars in Cancer Biology 2017, 43, 17–34.
Click here to remove instruction text...
2
3. Parks, S. K.; Pouyss gur, J. Semin. Cancer Bio. 2017, 43, 66–73.